A. Calpain
Calpain is a calcium-activated neutral protease, also known as CANP; EC 3.4.22.17. It is an intracellular cysteine protease which is ubiquitously expressed in mammalian tissues (Aoki et al., FEBS Letters 205:313-317, 1986). Calpain has been implicated in many degenerative diseases including, but not limited to, neurodegeneration (Alzheimer's disease, Huntington's disease, and Parkinson's disease), amyotrophy, stroke, motor neuron damage, acute central nervous system (CNS) injury, muscular dystrophy, bone resorption, platelet aggregation, and inflammation.
Mammalian calpain, including human calpain, is multimeric. It consists of two different subunits, which are a 30 kDa subunit and an 80 kDa subunit, and, therefore, is a heterodimer. There are two forms of calpain, calpain I (.mu.-calpain, .mu.CANP) and calpain II (m-calpain, mCANP), which differ in their sensitivities to the concentration of calcium necessary for activation. Calpain I requires only low micromolar concentrations of calcium for activation, whereas calpain II requires high micromolar or millimolar levels (Aoki et al. supra, and DeLuca et al., Biochim. Biophys. Acta 1216:81-83, 1993). The same 30 kDa subunit is common to both forms. The two human calpains differ in the sequences of the DNA encoding their 80 kDa subunit, sharing 62% homology. There is evidence that the 80 kDA subunit is inactive, but that it is autolyzed to a 76 kDa active form in the presence of calcium (Zimmerman et al., Biochem. Biophys. Acta., 1078:192-198, 1991).
R. Siman, in Neurotoxicity of Excitatory Amino Acids, A. Guidotti, ed., Raven Press , Ltd., New York (1990) reported upon the role of calpain I in excitatory amino acid (EAA) induced neurotoxicity, eventually leading to neuronal cell death. Siman advanced the proposition that calpain I activation is an early event in the neurodegenerative process and not just a secondary response to neuronal death. Siman further reported that only one highly selective blocker of calpain was available at that time--calpastatin. However, calpastatin is not readily taken up by cells, as it is a large globular protein of approximately 280 kDa. Siman also reported that protease inhibitors of broader specificity, including leupeptin, were unsuccessful in lowering EAA-induced protein breakdown in vivo. Leupeptin was ineffective presumably because it failed to enter the cells.
Iwamoto et al., Brain Research, 561:177-180 (1991), described that activation of calpain may be an important factor in the abnormal proteolysis underlying the accumulation of plaque and tangles in brain tissue from people who suffered Alzheimer-type dementia.
Saito et al., Proc. Natl. Acad. Sci. USA, 90:2628-2632 (1993) reported that synaptic loss and neuronal cell death correlate strongly with the degree of cognitive impairment in Alzheimer's disease. They also reported that calpain I was significantly activated in human postmortem brain from patients with Alzheimer's disease, and that the degree of activation correlated with those regions of the brain showing the greatest amount of degeneration. It was suggested that the influences of calpain activation may contribute to neurofibrillary pathology and abnormal amyloid precursor protein processing prior to causing synapse loss or cell death in the most vulnerable neuronal populations. Because of the association between calpain and nerve degeneration diseases, pharmacological modulation of the calpains by inhibitors merits consideration as a potential therapeutic strategy in such diseases, for example, in Alzheimer's disease.
Rami et al., Brain Research, 609:67-70 (1993) reported that both calpain inhibitor I and leupeptin protected neurons against ischemic and hypoxic damage resulting from ischemia induced by clamping both carotid arteries and lowering the arterial blood pressure of rats.
Lee et al., Proc. Natl. Acad. Sci. USA, 88:7233-7237 (1991) provide evidence that calcium-activated proteolysis is an important event in the process of post-ischemic cell death and they reported that inhibition of calcium-activated proteolysis by means of the proteolytic inhibitor leupeptin protected against the degeneration of vulnerable hippocampal neurons after ischemia. Leupeptin was selected because it was the only protease inhibitor that was previously shown to block a trauma-evoked calpain response in vivo (Seubert et al., Brain Res., 459:226-232, 1988) . The authors noted, however, that the therapeutic utility of modulating calcium-activated proteolysis will probably depend on the development of more permeable, potent and specific protease inhibitors.
As evident from the foregoing, specific inhibitors of calpain may provide a means of treating those neurodegenerative diseases in which calpain is implicated. Calpastatin offers limited utility due to its cell impermeability. Protease inhibitors of broader specificities may not function in vivo and/or may have undesirable side-effects. Thus, other calpain inhibitors must be identified, and a ready, convenient, safe source of calpain will promote the search for such inhibitors.
B. Calpain cDNA
Recombinant enzymatically active human calpain for testing for inhibitors offers the advantages of 1) being a considerably more convenient, readily available source of large amounts of enzyme 2) being easier to purify and 3) being free from the safety issues which must be addressed when the source is human tissues, especially human blood cells, i.e., potentially hazardous viruses. Native human calpain is currently isolated from human erythrocytes and can be purified to what the authors characterize as apparent homogeneity (Hatanaka et al., Biomed. Res., 4:381-388, 1983). However, aside from the obvious problems with source, the purification procedure can be quite tedious, due to the low levels of calpain relative to the amount of starting material. Furthermore, native calpain is isolated in the presence of an endogenous inhibitor (calpastatin) which must be separated during purification. A good source of large amounts of enzymatically active calpain would greatly enhance the search for calpain inhibitors by 1) increasing the availability of calpain for use in reproducible assays for calpain inhibitors and 2) by facilitating crystallization of the enzyme, thereby permitting the design of rational inhibitors. A recombinant system for production further facilitates the production of directed mutants to assist in structural studies. Therefore, a recombinant system for producing active calpain is needed.
The problem in producing enzymatically active calpain by recombinant means is that of expressing two different gene products (the 80 kDa subunit and 30 kDa subunit), getting proper processing and folding of the individual products, and obtaining the proper combination of the two products to produce enzymatically active molecules. As stated in the previous discussion, activated calpain has been implicated in the killing of neuronal cells. Unfortunately, then, any enzymatically active calpain produced in a recombinant system would be expected to be deleterious or lethal to that expression system. Any deleterious effects upon the expression system utilized would be expected to increase as more of the activated product is expressed. Notably, many mammalian cells produce an endogenous inhibitor of calpain, which may exert an important control over the activity of an otherwise lethal protease.
Aoki et al., supra, described the complete amino acid sequence of the 80 kDA subunit of human calpain I (.mu.CANP) which they deduced from the sequence of a cDNA clone of human calpain. The cDNA clone of human calpain was isolated from the cDNA library from human skeletal muscle using a cDNA for the large subunit of rabbit .mu.CANP as a probe. Expression of the cDNA is not reported.
Imajoh et al., Biochemistry, 27:8122-8128 (1988) described the isolation of a cDNA clone for the large subunit of human calpain II from a human skeletal muscle library probed with chicken CANP and rabbit .mu.CANP. It is reported that the deduced protein had essentially the same structural features as those described for .mu.CANP and chicken CANP. The amino acid sequence similarities of the human mCANP to human .mu.CANP and chicken CANP were reported as 62% and 66%, respectively. Expression of the cDNA is not described or suggested.
Ohno et al., Nucleic Acids Research, 14:5559 (1986) described the sequence of a cDNA coding for the small subunit (30 kDa) of human calcium activated protease isolated from a human spleen cDNA library. Comparisons with the reported amino acid sequences of rabbit and porcine sequences revealed only 3% differences.
DeLuca et al., supra., reported the molecular cloning and bacterial expression of cDNA for the rat calpain II (mCANP) 80 kDa subunit. The cDNA encodes a protein reportedly exhibiting 93% sequence identity with human calpain II, and 61% identity with human calpain I. Expression of the cDNA was in E. coli bacteria in a phagemid expression vector. Because the expressed product was insoluble and inactive after cell sonication, it could not be used to screen for calpain inhibitors.
C. Baculovirus Expression Systems
V. Luckow, Current Opinion in Biotechnology, 4:546-572 (1993) and Kidd et al, Applied Biochem. and Biotech., 42:137-159 (1993) recently reviewed baculovirus systems for the expression of human gene products and the use of baculoviruses as expression vectors, respectively. Luckow discussed the production of a number of different kinds of proteins, including enzymes. However, the production of only one proteolytic enzyme is mentioned, namely, the metalloprotease stromelysin. Unlike calpain, this enzyme is not multimeric.
Kidd et al., discussed the use of baculovirus-produced proteins for X-ray structural analysis and for assembly of subunits to form functional multisubunit molecules. A number of examples displayed the proper assembly of the subunits to produce functional molecules. Although the author broadly stated that baculovirus expression results in the structural integrity of the folded molecules and full biological function in virtually all cases, the assembly of subunits of dimeric or multimeric enzymes into a functional enzyme was not reported. Further, in other instances involving multisubunit molecules, i.e. Na,K,ATP-ase, the assembly of subunits was sometimes inefficient. (See DeTomaso et al., infra.)
Others have reported the expression of enzymes in the baculovirus system. Vernet et al., J. Biol. Chem., 27:16661-16666 (1990) described the secretion of a papain precursor from insect cells. Papain is a cysteine protease. The prepropapain gene was cloned into the transfer vector IpDC125 behind the polyhedron promoter. The recombinant construct was incorporated by homologous recombination into the genome of the polyhedrosis virus Autographa californica. An enzymatically inactive papain precursor was recovered from Spodoptera frugiperda Sf9 cells infected with the recombinant baculovirus. Proper processing of the papain precursor to produce an active enzyme did not occur in the infected cells.
Fertig et al., Cytotechnology, 11:67-75 (1993) described the production of pro-kallikrein, which is a precursor of kallikrein, a serine protease. Pro-kallikrein was produced in insect cells from Spodoptera frugiperda (Sf9) and Mamestra brassicae (IZD-Mb503) infected with a recombinant nuclear polyhedrosis virus Autographa californica (AcNPV), strain E2.To obtain an active enzyme, the pro-kallikrein produced was activated in vitro using trypsin.
Button et al., Gene, 133:75-81 (1993) described the production of the metalloproteinase GP63 of Leishmania major in a baculovirus-insect cell expression system. The enzyme was secreted from Spodoptera frugiperda (Sf9) cells infected with a recombinant nuclear polyhedrosis virus Autographa californica (AcNPV) as a latent protease which was subsequently activated to full proteinase activity by means of HgCl2 treatment.
Hirowatari et al., Arch. Virol., 133:349-356 (1993) described the expression of a polypeptide believed to exhibit two viral proteinase activities required for the processing of the viral precursor protein of hepatitis C virus (HCV). The polypeptide was expressed in the insect cell line Sf21 infected with a recombinant baculovirus. Baculovirus transfer vector pVL941 was utilized. The proteinase activities were inferred from the presence of a 70 kDa processed protein.
Although the production of enzymatically active multimeric proteases in the baculovirus system has not, to the inventors' knowledge, been reported, the baculovirus system has been used to express functional, multimeric enzymes other than proteases. DeTomaso et al., J. Biol. Chem., 268(2):1470-1478 (1993) describe the expression of functional, rat Na,K-ATPase using the baculovirus expression system. An expression system using insect cells was chosen because some insect cells have little or no levels of Na,K-ATPase. A baculovirus system was chosen since baculovirus-infected cells produce high levels of foreign protein. Sf-9 cells derived from Spodoptera frugiperda were utilized. The baculovirus was Autographa californica. However, because the activity of enzyme from insect cells was only 20-25% as great as that from dog kidney outer medulla, the authors concluded that a portion of the enzyme expressed was inactive.
Wen-Ji et al., J. Biol. Chem., 268(13):9675-9680 (1993) describe the expression of functional mammalian protein farnesyltransferase in a baculovirus system using SF9 cells. The specific activity of the expressed protein was 510 nM/mg/hr, which is stated to be essentially identical to that reported for the rat brain enzyme. It was noted, however, that the quantities of protein obtained from native tissue did not previously allow direct assay of the protein concentration, so this is the first time specific activity of the protein was determined using a standard protein assay.
There is no disclosure, or suggestion, of expressing an enzymatically active, multimeric, potentially lethal, protease such as calpain in any expression system. It was expected that expression of calpain I, in particular, would be difficult and would require the presence of an inhibitor, because calpain I is activated at extremely low levels of Ca++ that could be achieved during the infection cycle. Surprisingly, the present inventors unexpectedly found that enzymatically active calpain can be expressed in the baculovirus system, and in the absence of an inhibitor.